From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pf1-f170.google.com (mail-pf1-f170.google.com [209.85.210.170]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 0A6C43CB3D for ; Thu, 12 Dec 2019 20:57:23 -0500 (EST) Received: by mail-pf1-f170.google.com with SMTP id y14so559134pfm.13 for ; Thu, 12 Dec 2019 17:57:23 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=CndJAowMJInL8tn9fXQ0Bjot5xFUhuMQ3mivrKVBjhs=; b=Mg2xfxUVJdQ25CJS1CIg8Hr9hIjd1GGPXIs/NIAvZwEsLOkeo6bw4Ea9Z65fiUkHjA iGplnEgc7P8fulD4MInKfwNtoTWohHkoPEMRG5UmHRFZG8A69E+Txt2XpvVCC4ucFajN cFzG1C5LK1+opAiXFmn6hYXf1UEk/MoWgedqCfRDNPpuBhidXuIgL3CgceWQWYL2EmRk iMu6z0dMmriY/+m7UknBolo14ILP1LPt3qI/3EWLGe3inr5uIO9NL3qhLc/kXMFdYgiD AbVDgG1YGivBGvqsSHMOFzFn3Go+zDAIFKBHHZfSLQcISTnZyMW99K7kfxnbANoowxKf GZ9A== X-Gm-Message-State: APjAAAU5fePkqsqT51YxQ53zIKf7L9iuo1qFywKprxH3vq8nM4b75uPj aUx11w3RTHLs9zFzjh33Kt8= X-Google-Smtp-Source: APXvYqz5+9I2c9+BB9gbo/XUtG0zU4JO/dvc8qFiDsEuM+T5MJLIqwsMikJPqffpMgKBhGYk9TAq4w== X-Received: by 2002:a63:fe0a:: with SMTP id p10mr13699053pgh.96.1576202242950; Thu, 12 Dec 2019 17:57:22 -0800 (PST) Received: from [192.168.51.23] (184-23-135-132.dedicated.static.sonic.net. [184.23.135.132]) by smtp.gmail.com with ESMTPSA id i11sm6421887pjg.0.2019.12.12.17.57.21 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 12 Dec 2019 17:57:22 -0800 (PST) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 11.5 \(3445.9.1\)) From: Simon Barber In-Reply-To: <34a05f62-8dd0-9ea0-2192-1da5bfe6d843@gmail.com> Date: Thu, 12 Dec 2019 17:57:13 -0800 Cc: Dave Taht , Make-Wifi-fast , Johannes Berg , linux-wireless , Neal Cardwell , Netdev Content-Transfer-Encoding: quoted-printable Message-Id: <5DBB113A-E61A-4032-B877-18BE3ABBC616@superduper.net> References: <14cedbb9300f887fecc399ebcdb70c153955f876.camel@sipsolutions.net> <99748db5-7898-534b-d407-ed819f07f939@gmail.com> <22B5F072-630A-44BE-A0E5-BF814A6CB9B0@superduper.net> <34a05f62-8dd0-9ea0-2192-1da5bfe6d843@gmail.com> To: Eric Dumazet X-Mailer: Apple Mail (2.3445.9.1) Subject: Re: [Make-wifi-fast] debugging TCP stalls on high-speed wifi X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 13 Dec 2019 01:57:24 -0000 In my application this is a bridge or router (not TCP endpoint), and the = driver is doing GRO and NAPI polling. Also looking at using the = skb->fraglist to make the GRO code more effective and more transparent = by passing flags, short segments, etc through for perfect reconstruction = by TSO. Simon > On Dec 12, 2019, at 5:46 PM, Eric Dumazet = wrote: >=20 >=20 >=20 > On 12/12/19 4:59 PM, Simon Barber wrote: >> I=E2=80=99m currently adding ACK thinning to Linux=E2=80=99s GRO = code. Quite a simple addition given the way that code works. >>=20 >> Simon >>=20 >>=20 >=20 > Please don't. >=20 > 1) It will not help since many NIC do not use GRO. >=20 > 2) This does not help if you receive one ACK per NIC interrupt, which = is quite common. >=20 > 3) This breaks GRO transparency. >=20 > 4) TCP can implement this in a more effective/controlled way, > since the peer know a lot more flow characteristics. >=20 > Middle-box should not try to make TCP better, they usually break = things.